- From: Matt Rakow via GitHub <sysbot+gh@w3.org>
- Date: Sat, 23 Feb 2019 00:36:42 +0000
- To: public-pointer-events@w3.org
Ah my mistake - I had forgotten that I worked around this issue by using document.elementFromPoint on each unpacked move event to get the target element. The unpacked moves each have the same target as you say. I'm also thinking about drawing scenarios, in particular scenarios where the canvas might be small (checkbox, signature field) and it's highly likely that a coalesced move may fall outside of the canvas but include coalesced points from inside the canvas. After thinking some more - for this scenario I suppose pointer capture would avoid issues 1 & 2 assuming the author doesn't need the enter/leave events (or is OK using elementFromPoint if they do). Fiddle I was playing around with here: https://jsfiddle.net/cokgdx7v/ I don't think "really raw" is what I would want in this case either. The timing of the delivery of the coalesced event is already perfect - I don't desire to receive them any faster than once per frame because I'll only be drawing once per frame anyway. I just want to ensure that the quantity, coordinates, and targets of those coalesced events match what I would have received in a non-coalesced world. -- GitHub Notification of comment by ChumpChief Please view or discuss this issue at https://github.com/w3c/pointerevents/issues/278#issuecomment-466595443 using your GitHub account
Received on Saturday, 23 February 2019 00:36:44 UTC